dbstack

Index Categories

§ Sections

Browse by category.

Six rough buckets covering most database workloads. Pick a section to see every tool we track in it, ranked by stars.

  1. § 01 · Relational

    4 tools

    Relational databases.

    The default category for most applications. Relational databases trade some write scalability for transactional guarantees, joins, foreign keys, and a half-century of operational wisdom.

    Tracked: CockroachDB · PostgreSQL · MySQL · SQLite

    Read the section →
  2. § 02 · Postgres-based platforms

    3 tools

    Postgres-based platforms.

    Postgres is the wire protocol of modern databases. A whole layer of products — sharding extensions, serverless services, full-stack platforms — sits on top, offering Postgres compatibility with different operational profiles.

    Tracked: Supabase · Neon · Citus

    Read the section →
  3. § 03 · Analytical (OLAP)

    2 tools

    Analytical (OLAP) databases.

    When you need to scan billions of rows to compute aggregates rather than read or update individual records, column stores dominate. They trade single-row update speed for orders-of-magnitude faster scans and aggressive compression.

    Tracked: ClickHouse · DuckDB

    Read the section →
  4. § 04 · Key-value

    2 tools

    Key-value stores.

    KV stores trade query expressiveness for raw speed. They sit in front of databases as caches, handle session storage and rate limiting, and serve as lightweight queues and pub/sub brokers.

    Tracked: Redis · DragonflyDB

    Read the section →
  5. § 05 · Document

    1 tool

    Document databases.

    Document databases optimize for the case where each record is more or less self-contained. They trade relational joins for schema flexibility — useful when the application data model is fluid or when documents map naturally to API responses.

    Tracked: MongoDB

    Read the section →
  6. § 06 · Time-series

    1 tool

    Time-series databases.

    Time-series workloads (metrics, IoT, financial ticks) have predictable shapes — append-mostly, time-indexed, queried by time range. Purpose-built systems exploit this with partitioning, compression, and incremental rollups.

    Tracked: TimescaleDB

    Read the section →